Statistics Help Much Appreciated

#1
Hi, I'm revising for my upcoming exam and I have pastpapers but they don't give us the answers and I am unsure whether I have two questions right and I don't want to be learning the wrong answers, if someone could help I'd much appreciate it, Thankyou.

For 5.1 I thought the answer was B
and 5.2 I thought the answer was D

5.1. H1 in a chi-square test states:
a) The distribution of scores in two samples does not come from the same population
b) The distribution of frequencies in two samples does not come from the same
population
c) The order of scores in two samples does not occur randomly
d) The variance in true scores is larger than the variance in error scores
e) The difference between scores in two samples is due to the independent variable

5.2. A variance estimate is defined as:
a) The mean square of the true variation in scores divided by the mean square of the
error variation in scores
b) The difference between a particular score and the mean of the sample of scores from
which it is drawn, divided by the standard deviation
c) The standard deviation divided by the square root of the sample size
d) The sum of the squared deviation of the means, divided by the degrees of freedom
e) The average deviation of sample means from the mean of sample means
 
#2
There are many "chi square" test (which mean different things) and its not clear what H1 is here. Does it mean the alternative hypothesis (with Ho being the null hypothesis)?

I think D is correct for 5.2 although the wording is different than I have seen it. Variance is the sum of the squared differences between observed data and the mean divided by n or( n-1 in a sample). I have not seen it defined in terms of Df although there are lots of way to define terms in statistics :p
 

Dason

Ambassador to the humans
#3
I have not seen it defined in terms of Df although there are lots of way to define terms in statistics :p
It's probably slightly more appropriate if you think about it in a regression context. The degrees of freedom associated with the error is the number of observations minus the number of parameters in the model.

I'm not a fan of the wording of these questions though. I share noetsi's concerns about 5.1 and for 5.2 I feel like the answer should be "an estimate of the variance" ;)
 
#5
Yes they mean h1 as the alternative hypothesis and ho as the null, does that make the answer any clearer? thanks for the replies everyone :)
 
#6
Again there are a lot of chi square test which mean very different things. I am guessing that this refers to a pearson chi square test of a theoretical distribution (you are testing if the distribution you observe matches a theoretical one). If that is the case then b seems reasonable althoug the wording is really confusing to me. I think this would be a better way to state the alternative hypothesis in this case

(The hypothesis involves if the observed data matches a Poisson distribution0

H1: The number of arrivals per minute does not follow a Poisson distribution

http://courses.wcupa.edu/rbove/Berenson/10th ed CD-ROM topics/section12_5.pdf

But other chi square test would lead to very different alternate hypothesis and I really am not sure what I cited means the same thing as b in 5.1:p
 
Last edited:

Dason

Ambassador to the humans
#7
Again there are a lot of chi square test which mean very different things. I am guessing that this refers to a pearson chi square test of a theoretical distribution (you are testing if the distribution you observe matches a theoretical one).
If I had to guess the wording of the answers sound more appropriate for a chi-square test of independence.